Method and system for determining drivable road regions for safe navigation of an autonomous vehicle

Abstract

The present disclosure discloses method and a road region determination system for determining drivable road regions for safe navigation of an autonomous vehicle. The road region determination system receives real-time images of road in which autonomous vehicle is travelling, from image sensors. Each of real-time images of road is segmented into polygon regions and trend lines to obtain plurality of features using pre-trained road segmentation model. An orientation of the road is identified in real-time images as one of linear orientation and non-linear orientation based on slope measured between successive intermediate points distributed evenly on trend lines. The road region determination system manages redistribution of intermediate points on trend lines based on orientation of the road. Thereafter, paired points from intermediate points redistributed on the trend lines is identified, where paired points are connected using a horizontal line to determine drivable road regions for autonomous vehicle.

We claim: 
     
         1 . A method for determining drivable road regions for safe navigation of an autonomous vehicle, the method comprising:
 receiving, by a road region determination system, real-time images of a road in which an autonomous vehicle is travelling, from one or more image sensors, associated with the autonomous vehicle;   segmenting, by the road region determination system, each of the real-time images of the road into polygon regions and trend lines to obtain a plurality of features associated with the road using a pre-trained road segmentation model, wherein the road segmentation model is trained with a machine learning technique, using a plurality of training images marked with road features, polygon regions and trend lines;   identifying, by the road region determination system, orientation of the road in the real-time images to be one of, a linear orientation and a non-linear orientation based on a slope measured between successive intermediate points distributed evenly on the trend lines;   managing, by the road region determination system, redistribution of the intermediate points on the trend lines based on the orientation of the road; and   identifying, by the road region determination system, paired points from the intermediate points redistributed on the trend lines, wherein the paired points are connected using a horizontal line to determine the drivable road regions for the autonomous vehicle.   
     
     
         2 . The method as claimed in  claim 1 , wherein the polygon regions comprises an entire road polygon region, left road polygon region and right road polygon region and the trend lines comprise a left trend line on the left road polygon region and a right trend line on the right road polygon region. 
     
     
         3 . The method as claimed in  claim 1 , wherein the plurality of features associated with the road comprises type of road and colour of road. 
     
     
         4 . The method as claimed in  claim 1 , wherein identifying the orientation of the road to be one of, the linear orientation and the non-linear orientation comprises:
 distributing, by the road region determination system, the intermediate points evenly on a left trend line and a right trend line of the road based on predefined distance;   calculating, by the road region determination system, the slope between each successive intermediate point; and   identifying, by the road region determination system, the orientation of the road to be linear and non-linear when the slope between each successive intermediate point is within a predefined threshold range and beyond the predefined threshold range of slopes respectively.   
     
     
         5 . The method as claimed in  claim 1 , wherein managing the redistribution of the intermediate points on the trend lines comprises altering the distribution of the intermediate points by placing more intermediate points on curved region of the road, on identifying the orientation of the road to be non-linear. 
     
     
         6 . The method as claimed in  claim 1 , wherein managing the redistribution of the intermediate points on the trend lines comprises maintaining the evenly distribution of the intermediate points on identifying the orientation of the road to be linear. 
     
     
         7 . The method as claimed in  claim 1 , wherein identifying the paired points of the intermediate point comprises connecting each intermediate point on a left trend line with corresponding intermediate point on a right trend line with the horizontal line. 
     
     
         8 . The method as claimed in  claim 1 , wherein the road regions comprises left road boundary region, right road boundary region and angle of curvature of the road. 
     
     
         9 . The method as claimed in  claim 1  further comprising communicating the determined road regions to a control unit of the autonomous vehicle for triggering respective units in the autonomous vehicle based on the road regions for safe navigation. 
     
     
         10 . A road region determination system for determining drivable road regions for safe navigation of an autonomous vehicle, comprising:
 a processor; and   a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, causes the processor to:
 receive real-time images of a road in which an autonomous vehicle is travelling, from one or more image sensors, mounted in the autonomous vehicle; 
 segment each of the real-time images of the road into polygon regions and trend lines to obtain a plurality of features associated with the road using a pre-trained road segmentation model, wherein the road segmentation model is trained with a machine learning technique, using a plurality of training images marked with road features, polygon regions and trend lines; 
 identify orientation of the road in the real-time images to be one of, a linear orientation and a non-linear orientation based on a slope measured between successive intermediate points distributed evenly on the trend lines; 
 manage redistribution of the intermediate points on the trend lines based on the orientation of the road; and 
 identify paired points from the intermediate points redistributed on the trend lines, wherein the paired points are connected using a horizontal line to determine the drivable road regions for the autonomous vehicle. 
   
     
     
         11 . The road region determination system as claimed in  claim 10 , wherein the polygon regions comprises an entire road polygon region, left road polygon region and right road polygon region and the trend lines comprise a left trend line on the left road polygon region and a right trend line on the right road polygon region. 
     
     
         12 . The road region determination system as claimed in  claim 10 , wherein the plurality of features associated with the road comprises type of road and colour of road. 
     
     
         13 . The road region determination system as claimed in  claim 10 , wherein the processor identifies the orientation of the road to be one of, the linear orientation and the non-linear orientation by:
 distributing the intermediate points evenly on a left trend line and a right trend line of the road based on predefined distance;   calculating the slope between each successive intermediate point; and   identifying the orientation of the road to be linear and non-linear when the slope between each successive intermediate point is within a predefined threshold range and beyond the predefined threshold range of slopes respectively.   
     
     
         14 . The road region determination system as claimed in  claim 10 , wherein the processor manages the redistribution of the intermediate points on the trend lines by altering the distribution of the intermediate points by placing more intermediate points on curved region of the road, on identifying the orientation of the road to be non-linear. 
     
     
         15 . The road region determination system as claimed in  claim 10 , wherein the processor manages the redistribution of the intermediate points on the trend lines by maintaining the evenly distribution of the intermediate points on identifying the orientation of the road to be linear. 
     
     
         16 . The road region determination system as claimed in  claim 10 , wherein the processor identifies the paired points of the intermediate point by connecting each intermediate point on a left trend line with corresponding intermediate point on a right trend line with the horizontal line. 
     
     
         17 . The road region determination system as claimed in  claim 10 , wherein the road regions comprises left road boundary region, right road boundary region and angle of curvature of the road. 
     
     
         18 . The road region determination system as claimed in  claim 10 , wherein the processor is configured to communicate the determined road regions to a control unit of the autonomous vehicle for triggering respective units in the autonomous vehicle based on the road regions for safe navigation.